chem-nmr-predict

To predict a 1H NMR spectrum from SMILES, you provide a valid SMILES string and an optional spectrometer field strength in MHz to generate .xy spectra and signal tables.

Generating reference NMR spectra for deconvolution is best done by predicting 1H NMR spectra from SMILES to produce .xy files and signal tables compatible with downstream analysis workflows.

Yes, you can perform multi-compound batch predictions by providing multiple SMILES strings to generate predicted 1H NMR spectra and signal tables for each structure.

Yes, NMR spectrum prediction supports different field strengths by allowing you to optionally specify the spectrometer frequency in MHz alongside your valid SMILES input.

Chemical shifts and coupling constants are estimated using the NMRdb.org SPINUS neural-network predictions, combined with nmrsim quantum-mechanical spin simulation for peak generation.

If SPINUS or nmrsim fails during spectrum prediction, the workflow applies predefined fallback mechanisms to handle the failure modes and continue processing your SMILES input.